regression in jammy kernel 5.15 for module ath9k / same issue with mt76 - additional bug report to be created?

Asked by Peter Langner

facing issue (after migrating from 20.04 to 22.04) with module mt76 causing unstable wifi - same as already described in bug report: regression in jammy kernel 5.15 for module ath9k.

[ 28.086265] ================================================================================
[ 28.086284] UBSAN: invalid-load in /build/linux-WD899k/linux-5.15.0/net/mac80211/status.c:1164:21
[ 28.086294] load of value 255 is not a valid value for type '_Bool'
[ 28.086301] CPU: 0 PID: 114 Comm: kworker/u8:2 Not tainted 5.15.0-30-generic #31-Ubuntu
[ 28.086309] Hardware name: Intel(R) Client Systems NUC11ATKC4/NUC11ATBC4, BIOS ATJSLCPX.0029.2021.1117.2104 11/17/2021
[ 28.086314] Workqueue: mt76 mt76u_tx_status_data [mt76_usb]
[ 28.086335] Call Trace:
[ 28.086340] <TASK>
[ 28.086346] show_stack+0x52/0x58
[ 28.086360] dump_stack_lvl+0x4a/0x5f
[ 28.086372] dump_stack+0x10/0x12
[ 28.086378] ubsan_epilogue+0x9/0x45
[ 28.086385] __ubsan_handle_load_invalid_value.cold+0x44/0x49
[ 28.086392] ? usb_control_msg+0xfe/0x140
[ 28.086401] ieee80211_tx_status_ext.cold+0x4e/0x5f [mac80211]
[ 28.086568] mt76_tx_status_unlock+0x10c/0x160 [mt76]
[ 28.086590] mt76x02_send_tx_status+0x168/0x3f0 [mt76x02_lib]
[ 28.086611] mt76x02_tx_status_data+0x47/0x70 [mt76x02_lib]
[ 28.086628] mt76u_tx_status_data+0x64/0xc0 [mt76_usb]
[ 28.086639] process_one_work+0x228/0x3d0
[ 28.086650] worker_thread+0x53/0x410
[ 28.086659] ? process_one_work+0x3d0/0x3d0
[ 28.086667] kthread+0x127/0x150
[ 28.086675] ? set_kthread_struct+0x50/0x50
[ 28.086683] ret_from_fork+0x1f/0x30
[ 28.086694] </TASK>
[ 28.086697] ================================================================================

New bug report to be created or should it be linked to the existing one? Thx.

Question information

Language:
English Edit question
Status:
Needs information
For:
Ubuntu Edit question
Assignee:
No assignee Edit question
Last query:
Last reply:
Revision history for this message
Manfred Hampl (m-hampl) said :
#1

I see that you already have added your information to Bug #1968631
I suggest to wait for a reaction on the bug report by the developers. If they say that your case it too different, and that you should create your own bug report, then do so. For the time being I suggest to leave it as it is.

Revision history for this message
Manfred Hampl (m-hampl) said :
#2

There is a new version of the kernel packages in jammy-proposed, 5.15.0-32.33
The change log shows modifications related to ath9k and mt76.
You might consider trying that version to check whether the problem is solved in the newer version.

Revision history for this message
Peter Langner (posten11) said :
#3

Many thanks.
Installed 5.15.0-35-generic from deb https://ppa.launchpadcontent.net/canonical-kernel-team/proposed/ubuntu jammy main

but problem still persist

 ================================================================================
Jun 2 23:34:37 warbird kernel: [ 90.469915] UBSAN: invalid-load in /build/linux-7F6rHG/linux-5.15.0/net/mac80211/status.c:1164:21
Jun 2 23:34:37 warbird kernel: [ 90.469924] load of value 255 is not a valid value for type '_Bool'
Jun 2 23:34:37 warbird kernel: [ 90.469931] CPU: 2 PID: 9 Comm: kworker/u8:0 Not tainted 5.15.0-35-generic #36-Ubuntu
Jun 2 23:34:37 warbird kernel: [ 90.469939] Hardware name: Intel(R) Client Systems NUC11ATKC4/NUC11ATBC4, BIOS ATJSLCPX.0029.2021.1117.2104 11/17/2021
Jun 2 23:34:37 warbird kernel: [ 90.469944] Workqueue: mt76 mt76u_tx_status_data [mt76_usb]
Jun 2 23:34:37 warbird kernel: [ 90.469966] Call Trace:
Jun 2 23:34:37 warbird kernel: [ 90.469971] <TASK>
Jun 2 23:34:37 warbird kernel: [ 90.469977] show_stack+0x52/0x58
Jun 2 23:34:37 warbird kernel: [ 90.469987] dump_stack_lvl+0x4a/0x5f
Jun 2 23:34:37 warbird kernel: [ 90.470000] dump_stack+0x10/0x12
Jun 2 23:34:37 warbird kernel: [ 90.470007] ubsan_epilogue+0x9/0x45
Jun 2 23:34:37 warbird kernel: [ 90.470014] __ubsan_handle_load_invalid_value.cold+0x44/0x49
Jun 2 23:34:37 warbird kernel: [ 90.470022] ? usb_control_msg+0xfe/0x140
Jun 2 23:34:37 warbird kernel: [ 90.470031] ieee80211_tx_status_ext.cold+0x4e/0x5f [mac80211]
Jun 2 23:34:37 warbird kernel: [ 90.470197] mt76_tx_status_unlock+0x10c/0x160 [mt76]
Jun 2 23:34:37 warbird kernel: [ 90.470220] mt76x02_send_tx_status+0x168/0x3f0 [mt76x02_lib]
Jun 2 23:34:37 warbird kernel: [ 90.470241] mt76x02_tx_status_data+0x47/0x70 [mt76x02_lib]
Jun 2 23:34:37 warbird kernel: [ 90.470258] mt76u_tx_status_data+0x64/0xc0 [mt76_usb]
Jun 2 23:34:37 warbird kernel: [ 90.470268] process_one_work+0x228/0x3d0
Jun 2 23:34:37 warbird kernel: [ 90.470280] worker_thread+0x53/0x410
Jun 2 23:34:37 warbird kernel: [ 90.470288] ? process_one_work+0x3d0/0x3d0
Jun 2 23:34:37 warbird kernel: [ 90.470296] kthread+0x127/0x150
Jun 2 23:34:37 warbird kernel: [ 90.470304] ? set_kthread_struct+0x50/0x50
Jun 2 23:34:37 warbird kernel: [ 90.470312] ret_from_fork+0x1f/0x30
Jun 2 23:34:37 warbird kernel: [ 90.470323] </TASK>
Jun 2 23:34:37 warbird kernel: [ 90.470326] ================================================================================

 On Ubuntu 20.04.4 LTS with same hardware config wifi was very stable.

Would there be a possibility to backport to 5.4.0-107-generic?

Many thanks!

Revision history for this message
Manfred Hampl (m-hampl) said :
#4

In the change log entries for the past kernel updates, I see changes for ath9k and mt76 in each of them.
Apparently the kernel developers are still working on a fix. Hopefully the problem will be solved soon.

"Would there be a possibility to backport to 5.4.0-107-generic?"
What do you want to backport from where to where?

Can you help with this problem?

Provide an answer of your own, or ask Peter Langner for more information if necessary.

To post a message you must log in.